Extreme and extensible applications Gigaspaces XAP and VMware vFabric GemFire

Tools to consider when you have extreme transactional or data intensive tasks
Written by Dan Kusnetzky, Contributor

From time to time, I receive messages from suppliers of technology that crosses several boundaries in the Kusnetzky Group Model of virtualization technology (see Sorting out the different layers of virtualization for more information on the model) making it hard to compare it to technology offered by others. GigaSpaces XAP and competitors such as VMware vFabric GemFire (formerly from GemStone are good examples of technology that contain features of application virtualization, virtual processing software and management of virtualized environments targeting a specific application area.

In the case of this type of software the suppliers' goal is to provide an environment that offers the following benefits:

  • Allows development of transaction or data intensive application using standard tools
  • Allows applications to flexibly scale as workloads require. The scaling could consume computing resources in a company's own datacenter(s) or flow out into the cloud as needed.
  • Quite often they also offer a different take on data management allowing massive amounts of data to be manipulated using a non-relational/non-SQL-based environment

The suppliers of this technology are finding customers needing "extreme" transactional levels, helping them achieve their goals and then publicizing it broadly in the hopes of interesting others. I've seen references to social gaming sites, risk management or other technical computing applications and other intensive workloads.

Are these tools broadly applicable? I don't think so. Many workloads either are built using architectures or have interdependent data that don't lend themselves well to this distributed/parallel approach to computing.

When are where the the applications can work using this approach, however, these tools can be a Godsend. In this case, without these tools developers would have to create their own tools that combine the needed features and, of course, support them as long as that workload is required. Few organizations want to do that type of development and support in today's market.

If your organization needs to boost workload performance and has applications that can be spread over a computational or data grid/cluster/pick your favorite buzzword configuration, these tools could save the day.  Over the years I've spoken with representatives of both Gigaspaces and GemStone/now VMware and thought that they had interesting technology worthy of consideration.

Editorial standards